AI Engineer
GetSetYo | Bengaluru, Karnataka, India | Today
full-time | on-site | entry | 1–8 years | bachelor in Engineering
skills: python, java, llm, rag, vector search, api development, backend integration, deployment, monitoring, evaluation systems, observability, debugging, prompt design, tool calling, agent frameworks, mcp
Responsibilities
- Build and own production-grade AI features across the stack, from experimentation and prototyping to backend integration, deployment, monitoring, and iterative improvement.
- Design and implement agentic workflows for real user problems combining LLM reasoning, retrieval, tool use, business rules, and backend APIs into reliable multi-step systems.
- Build and optimize RAG and search systems: document ingestion, chunking strategies, embedding pipelines, vector indexes, hybrid retrieval, reranking, and citation/grounding flows.
- Integrate models with internal and external systems through tool calling, APIs, and, where relevant, MCP-compatible interfaces, so models can safely access the right context and take useful actions.
- Drive context engineering for AI products: decide what memory, instructions, retrieved context, tool outputs, and interaction history should be passed to the model at each step for maximum quality and efficiency.
- Build evaluation systems for prompts, agents, and retrieval quality including benchmark datasets, golden test cases, automated regression checks, and human-in-the-loop review workflows.
- Establish observability and debugging for AI pipelines: traces, tool execution logs, latency/cost tracking, hallucination analysis, and failure-mode investigation.
- Help define engineering standards for AI systems across security, guardrails, versioning, rollback, experimentation, and cost-performance tradeoffs.
- 1-8 years of total software engineering experience, including at least 1 year building and shipping AI/ML or LLM-powered products in production
- An engineering degree from a top-ranked college.
- Strong engineering foundation in Python or Java, with the ability to build reliable backend services, APIs, evaluation pipelines, and developer tooling around AI systems.
- Hands-on experience with LLM application patterns such as RAG, tool/function calling, structured output generation, vector search, reranking, and agentic workflows.
- Familiarity with agent frameworks and orchestration patterns, including multi-step workflows, planner/executor patterns, tool routing, and guardrails.
- Working knowledge of MCP (Model Context Protocol) or similar patterns for connecting models to internal tools, data sources, and external systems.
- Strong understanding of context engineering, prompt design, and how to manage instructions, conversation state, tools, memory, and retrieved context for consistent model behaviour.
- Experience with evaluation and observability for AI systems: offline evals, online metrics, regression testing, trace inspection, cost/latency monitoring, and failure analysis.
- Comfortable working in a fast-paced startup where you can own problem statements end-to-end from prototype to production rollout.
- Must have experience using AI-native developer tools such as Claude Code/coding agents / AI-assisted workflows to accelerate delivery.
Get new builder jobs daily: